In LTLA/scRNAseq: Collection of Public Single-Cell RNA-Seq Datasets

knitr::opts_chunk$set(error=FALSE, warning=FALSE, message=FALSE)

Introduction

The r Biocpkg("scRNAseq") package provides convenient access to several publicly available single-cell datasets in the form of SingleCellExperiment objects. We do all of the necessary data munging for each dataset beforehand, so that users can obtain a SingleCellExperiment for immediate use in further analyses. To enable discovery, each dataset is decorated with metadata such as the study title/abstract, the species involved, the number of cells, etc. Users can also contribute their own published datasets to enable re-use by the wider Bioconductor community.

Finding datasets

The surveyDatasets() function will show all available datasets along with their metadata. This can be used to discover interesting datasets for further analysis.

library(scRNAseq)
all.ds <- surveyDatasets()
all.ds

Users can also search on the metadata text using the searchDatasets() function. This accepts both simple text queries as well as more complicated expressions involving boolean operations.

# Find all datasets involving pancreas.
searchDatasets("pancreas")[,c("name", "title")]

# Find all mm10 datasets involving pancreas or neurons.
searchDatasets(
   defineTextQuery("GRCm38", field="genome") &
   (defineTextQuery("neuro%", partial=TRUE) | 
    defineTextQuery("pancrea%", partial=TRUE))
)[,c("name", "title")]

Keep in mind that the search results are not guaranteed to be reproducible - more datasets may be added over time, and existing datasets may be updated with new versions. Once a dataset of interest is identified, users should explicitly list the name and version of the dataset in their scripts to ensure reproducibility.

Loading a dataset

The fetchDataset() function will download a particular dataset, returning it as a SingleCellExperiment:

sce <- fetchDataset("zeisel-brain-2015", "2023-12-14")
sce

For studies that generate multiple datasets, the dataset of interest must be explicitly requested via the path= argument:

sce <- fetchDataset("baron-pancreas-2016", "2023-12-14", path="human")
sce

By default, array data is loaded as a file-backed DelayedArray from the r Biocpkg("HDF5Array") package. Setting realize.assays=TRUE and/or realize.reduced.dims=TRUE will coerce these to more conventional in-memory representations like ordinary arrays or dgCMatrix objects.

assay(sce)
sce <- fetchDataset("baron-pancreas-2016", "2023-12-14", path="human", realize.assays=TRUE)
class(assay(sce))

Users can also fetch the metadata associated with each dataset:

str(fetchMetadata("zeisel-brain-2015", "2023-12-14"))

Adding new datasets

Want to contribute your own dataset to this package? It's easy! Just follow these simple steps for instant fame and prestige.

1. Format your dataset as a SummarizedExperiment or SingleCellExperiment. Let's just make up something here.

   r library(SingleCellExperiment) sce <- SingleCellExperiment(list(counts=matrix(rpois(1000, lambda=1), 100, 10))) rownames(sce) <- sprintf("GENE_%i", seq_len(nrow(sce))) colnames(sce) <- head(LETTERS, 10)

2. Assemble the metadata for your dataset. This should be a list structured as specified in the Bioconductor metadata schema Check out some examples from fetchMetadata() - note that the application.takane property will be automatically added later, and so can be omitted from the list that you create.

   r meta <- list( title="My dataset", description="This is my dataset", taxonomy_id="10090", genome="GRCh38", sources=list( list(provider="GEO", id="GSE12345"), list(provider="PubMed", id="1234567") ), maintainer_name="Chihaya Kisaragi", maintainer_email="[e-mail]" )

3. Save your SummarizedExperiment (or whatever object contains your dataset) to disk with saveDataset(). This saves the dataset into a "staging directory" using language-agnostic file formats - check out the alabaster framework for more details. In more complex cases involving multiple datasets, users may save each dataset into a subdirectory of the staging directory.

   ```r

   Simple case: you only have one dataset to upload.

   staging <- tempfile() saveDataset(sce, staging, meta) list.files(staging, recursive=TRUE)

   Complex case: you have multiple datasets to upload.

   staging <- tempfile() dir.create(staging) saveDataset(sce, file.path(staging, "foo"), meta) saveDataset(sce, file.path(staging, "bar"), meta) # etc. ```

You can check that everything was correctly saved by reloading the on-disk data into the R session for inspection:

```r
alabaster.base::readObject(file.path(staging, "foo"))
```

1. Open a pull request (PR) for the addition of a new dataset. You will need to provide a few things here:
2. The name of your dataset. This typically follows the format of {NAME}-{SYSTEM}-{YEAR}, where NAME is the last name of the first author of the study, SYSTEM is the biological system (e.g., tissue, cell types) being studied, and YEAR is the year of publication for the dataset.
3. The version of your dataset. This is usually just the current date... or whenever you started putting together the dataset for upload. The exact date doesn't really matter as long as we can establish a timeline for later versions.

4. An Rmarkdown file containing the code used to assemble the dataset. This should be added to the scripts/ directory of this package, in order to provide some record of how the dataset was created.

5. Wait for us to grant temporary upload permissions to your GitHub account.

6. Upload your staging directory to gypsum backend with gypsum::uploadDirectory(). On the first call to this function, it will automatically prompt you to log into GitHub so that the backend can authenticate you. If you are on a system without browser access (e.g., most computing clusters), a token can be manually supplied via gypsum::setAccessToken().

   r gypsum::uploadDirectory(staging, "scRNAseq", "my_dataset_name", "my_version")

You can check that everything was successfully uploaded by calling fetchDataset() with the same name and version:

```r
fetchDataset("my_dataset_name", "my_version")
```

If you realized you made a mistake, no worries. Use the following call to clear the erroneous dataset, and try again:

```r
gypsum::rejectProbation("scRNAseq", "my_dataset_name", "my_version")
```

1. Comment on the PR to notify us that the dataset has finished uploading and you're happy with it. We'll review it and make sure everything's in order. If some fixes are required, we'll just clear the dataset so that you can upload a new version with the necessary changes. Otherwise, we'll approve the dataset. Note that once a version of a dataset is approved, no further changes can be made to that version; you'll have to upload a new version if you want to modify something.

Session information {-}

sessionInfo()

LTLA/scRNAseq documentation built on April 14, 2024, 7:13 a.m.